I just won't allow myself to get over-hyped until I know the official specs and can seem some actual performance results.
The problem is in the word "true"
it would be like arguing over whether or not the Pentium 4 was the first "true" 2GHz or 3GHz CPU given its radically different pipeline design and lower IPC wouldn't provide 2 or 3GHz performance relative to architectures we had been used to with the Pentium 3 and Athlon.
what we ultimately will need to focus on is the real world performance
I'm pretty sure I understand why I see people like RussianSensation make comments along the lines of Bulldozer not being "true" 8 core, and it largely seems to be stemmed from overcompensation against those who are ignorant of the architecture change who make comments along the lines of "ZOMG!! 3.8-4.2GHz ON 8 CORES, AMD CANT LOSE!!!"
When the sobering reality is that the other specs suggest a $320 price point and a 125W TDP which don't add up to a Sandy Bridge slayer considering that price point matches the 2600K (why not price it higher if its truly a faster/better product?) and the TDP exceeds it (2600K is only 95W stock). The best case guess from that information is that AMD's newest/best 8 core CPU will only be slightly better, albeit not hands down better, than Intel's newest/best 4 core.
Although I hope I'm wrong, I just won't allow myself to get over-hyped until I know the official specs and can seem some actual performance results.
Because quad cores are typically the CPUs with the lowest amount of cores that still run in high clock frequencies and have a large amount of cache.
I've seen enough already to be reasonably certain that these chips will return AMD to competitiveness across the price spectrum. However, I wholeheartedly agree with those of you who recognize that core counts and clock speeds are too variable across different architectures to read much from them alone - I'll wait for benchmarks, and see if the Bulldozer motherboards offer useful features (I still don't know of any front USB 3.0 header-enabled boards), as well as their pricing.
Again, BD is a 8 core CPU - it has 8 physical integer cores.
Performance or lack of it doesn't change that.
About pricing, I have no idea, except the only thing you can dsay based on that is speculation/guessing - otherwise how could the E6600 beat an AMD FX62 or an Athlon X2 5000+ if it was cheaper than both? Or how could the 4870 beat a GTX260 if it was $100 cheaper?
I'm in the same boat, I would be shocked if Bulldozer wasn't an obvious upgrade for my heavily threaded workloads, but for my gaming I have serious doubts about it coming close to making me want to give up my 4.8GHz 2600K rig completely (of which I'm actually considering an upgrade to Z68 in order to get SSD caching, my Steam folder alone is several hundred gigs in size)
And I'm not saying it does change anything, but you have to understand that Bulldozer is reintroducing that classic situation of misleading specs. Just like When the Pentium 4 came out and I'd see people react along the lines of, "HOLY CRAP, 1.5GHz!?!?!!!? The Athlon and Pentium 3 are only 1GHz!! That means the P4 is 50% faster!!!!" I see the same thing going on with cores now and people ogling over Bulldozer's 8 cores as if nothing could possibly be disappointing about it because, hey, it has EIGHT FREAKING CORES MAN!!! That's 50-100% moar cores than what you can get from Intel!!!
Because they were more efficient per amount of silicon used, so regardless of how the chips performed they could be priced lower because you could get more chips per wafer.
What's more is that the E6600 wasn't the flagship part, the appropriately far more expensive E6700 was. And in the case of the 4870, that was ATI's flagship part and couldn't hope to compete solely on brute performance with nVidia's flagship GTX280, so they started a price war which they could afford because of their far more gaming-efficient GPU design.
I suppose it is possible that Bulldozer is going to have grand slam performance and that AMD is pricing them aggressively to try and maximize adoption rates, but if that is true then that just means Intel would likely slash Sandy Bridge 1155 prices (which are already competitively low to begin with) to match. And that would lead to the question of why wouldn't AMD just bump up pricing a tier or two if their top end BD is hands-down faster and their lower end BD chips can perform at a higher priced Intel level instead of forcing Intel down?
It is a trade.
What you also need to consider is the cost, die size and power consumption of that non-shared architecture versus the shared one.
Additional 80% is an average, not best case or worst case.
I'm pretty sure I understand why I see people like RussianSensation make comments along the lines of Bulldozer not being "true" 8 core, and it largely seems to be stemmed from overcompensation against those who are ignorant of the architecture change who make comments along the lines of "ZOMG!! 3.8-4.2GHz ON 8 CORES, AMD CANT LOSE!!!"
Again, BD is a 8 core CPU - it has 8 physical integer cores.
Performance or lack of it doesn't change that.
amd flagship was hd 4870 x2, and its still doesn't change the fact that amd have $100 cheaper card while have better performance and actually make amd gain market share from nvdia
I suppose it is possible that Bulldozer is going to have grand slam performance and that AMD is pricing them aggressively to try and maximize adoption rates, but if that is true then that just means Intel would likely slash Sandy Bridge 1155 prices (which are already competitively low to begin with) to match. And that would lead to the question of why wouldn't AMD just bump up pricing a tier or two if their top end BD is hands-down faster and their lower end BD chips can perform at a higher priced Intel level instead of forcing Intel down?
For example, lets say that this $320 price tag does indeed mean that the 8 core model competes with the i7-2600K, maybe a bit slower at threads 1-4 but faster at threads 5-8 - you say it is a disappointment, but I see AMD moving from Core 2 Quad performance straight into SB territory - quite a big jump.
Phenom currently uses CMP, or chip-level multiprocessing (i.e. just cramming multiple cores on a single chip). A single Bulldozer module (i.e., 2 cores) would only have 80% of the performance of 2 fully fledged cores in CMP design, all things being equal. What this means is that without any other changes a 6-Core Bulldozer CPU is only 80% as fast as a 6-core Phenom II. This means AMD will have to compensate significantly with clock speed, IPC improvements and well more cores (hence the 8-core BD spec at $320).
I am just afraid some are misinterpreting this the opposite way -- i.e., "Bulldozer is like an 6-Core Phenom but even more efficient". In fact, what AMD slides are showing is the opposite: "If a 6-Core BD was 25% more efficient than Phenom II X6 per clock, it would only be as fast as an Phenom II X6 at the same clock speed."
What this would mean then is if the 80% efficiency is taken at face value from AMD's internal slides, then for BD to match Nehalem, it would need to be a whopping 75% more efficient per clock cycle than a Phenom II was (i.e., we need to go from 80% to 100% to match Phenom II and then another 40% to get to Nehalem over Phenom II --> 0.8 --> 1.4x or a 75% increase).
Phenom currently uses CMP, or chip-level multiprocessing (i.e. just cramming multiple cores on a single chip). A single Bulldozer module (i.e., 2 cores) would only have 80% of the performance of 2 fully fledged cores in CMP design, all things being equal. What this means is that without any other changes a 6-Core Bulldozer CPU is only 80% as fast as a 6-core Phenom II. This means AMD will have to compensate significantly with clock speed, IPC improvements and well more cores (hence the 8-core BD spec at $320).
I am just afraid some are misinterpreting this the opposite way -- i.e., "Bulldozer is like an 6-Core Phenom but even more efficient". In fact, what AMD slides are showing is the opposite: "If a 6-Core BD was 25% more efficient than Phenom II X6 per clock, it would only be as fast as an Phenom II X6 at the same clock speed."
What this would mean then is if the 80% efficiency is taken at face value from AMD's internal slides, then for BD to match Nehalem, it would need to be a whopping 75% more efficient per clock cycle than a Phenom II was (i.e., we need to go from 80% to 100% to match Phenom II and then another 40% to get to Nehalem over Phenom II --> 0.8 --> 1.4x or a 75% increase).
And why would Intel slash their price?
It isn't as if AMD could supply all the demand and so Intel could prefer to keep their margins even at a cost of a few percent market share.
Or if we prefer, why would AMD offer their processors at a higher price? Intel could simply slash their processors even more and/or introduce higher clocked parts.
It is possible to spin it whatever way we feel like it until performance numbers are out.
For example, lets say that this $320 price tag does indeed mean that the 8 core model competes with the i7-2600K, maybe a bit slower at threads 1-4 but faster at threads 5-8 - you say it is a disappointment, but I see AMD moving from Core 2 Quad performance straight into SB territory - quite a big jump.
here's the thing though, if the 8 core model is only a tad slower than the 2600 at threads 1-4, then you can pretty much say that the 4 core model is only a tad slower than the 8 core model at threads 1-4 as well. That is, if CMT works as well as AMD claims it should.
Not really.
The 4 core is 2 modules, so it would only be a tad slower at threads 1-2, but could be slower at threads 3-4 depending on how much the shared resources would bottleneck it, and wouldn't be able to do threads 5-8 due to no HT.
On the other hand, while the 4 core BD might be a bit slower at threads 1-2 against the core i3 it would be faster on threads 3-4.
If BD loses in singlethreaded performance, then what's the point?
Yeah so what you can add more cores. But what matters is how fast the Cpu is clock for clock compared to everything else.
Single thread performance is determined by frequency and the instruction/clock reached in that application. A clock for clock comparison is completely bogus since they are not created to run at the same frequency, are not sold at the same frequency and are not positionned against the same frequency.
Yes, we only have to look back to the P4 days to see how individual parameters alone, like clock frequency, or clock-for-clock IPC, can be misleading.
Its not that simple. If the next version of Photoshop starts using a certain extension that Intel has that AMD doesn't, the performance in the current version isn't going to be relevant to you. So if you see Intel with a huge lead in Vegas or something, you might want to still take that into consideration.
At the same time if Photoshop zooming/panning performance in Intel is way better than AMD in current versions, it could become completely irrelevant in a future version which uses GPU acceleration for these things. Same goes for video encoding, folding, and a bunch of other tasks that might not be CPU dependent at all in the future. I'm assuming AMD is planning on doing something with Llano's GPU to fight Quicksync.
Good point, but then that won't matter that much because presumably peformance reviews will be using old software because it's too difficult to use updated software. Of course, that may change depending on who has the performance lead...
But anyway to your main point, yes i agree, this is why performance on current software is so important rather than generations old software. That's supposed to be the purpose of tech review sites like Anandtech, etc.; to inform would be consumers of relevant information. Suggesting it's just too hard to keep data up to date is a sorry excuse.
Notice its anandtech.com, not anandtech.org.
Informing consumers is a means to an end, the end being profit Let's not get overzealous in our statements of purpose.
No different than amdzone.com.
Of course this would be assuming that the BD isn't improving over the phenom II - it is more likely that the 4 core BD beats the phenom X4, shared resources (which are larger to start with) or not.
That statement is so wrong in so many ways...
Clock for clock matters as much counting the eggs before you create vegatable soup.
Single thread performance is determined by frequency and the instruction/clock reached in that application. A clock for clock comparison is completely bogus since they are not created to run at the same frequency, are not sold at the same frequency and are not positionned against the same frequency.
There is no question a 4 core BD will beat a 4-core Phenom II but I just pointed out that it will need significant increases in IPC and clockspeed since out of the gate the shared module design is 25% slower.... With TurboMode I bet BD will run at 3.8-4.0ghz out of the box. But I am just saying it has a mountain to climb even to match a Nehalem in IPC in certain tasks. So I don't see at all how it can match SB in tasks that only require 4 threads (which let's face it for the most part are 90% of programs).
No it isn't bogus. Actually he is completely right. Since Intel has the manufacturing advantage, it is unlikely that BD will be able to overclock beyond 4.7ghz-4.8ghz that SB can already do now. We are enthusiasts here so for most of us stock performance out of the box is meaningless. Your argument for IPC being irrelevant would have made a lot more sense if we were in the era of Athlon 64 vs. Pentium-D or Athlon XP+ vs. P4 since the frequency ceiling was completely different during those generations. With SB, not only do you have the highest frequency ceiling, but you also have the leading IPC in the industry. So that means there is only 1 way to beat such a processor - more cores. This is exactly why BD will need 8 to compete with the 2600k.
It's reasonable to predict that for BD and SB the clock speeds are going to be fairly close at max overclocks. Therefore, clock for clock performance is ESPECIALLY important for this generation, just like it was critical during the Phenom II vs. Nehalem/Lynnfield era (i.e., both the Phenom II and Nehalem / Lynnfield pretty much topped out at ~ 4.0-4.2ghz, which made IPC the most relevant metric for most tasks). In fact, the main reason Phenom II lost to Core i generation was because of it lacking IPC performance.